Algolia

Beyond the search box, how digital shopping got smarter with an API-first approach
facebooklinkedintwittermail

Just as the pre-Internet man or woman in the street had “only” stores and malls to go to, the pre-recommendation engine e-commerce shopper has “only” the search box … today we have an ability to search smarter with API-driven machine intelligence designed to fill our carts and warm our hearts.

We used to go the shops, but those days have now passed. Okay, in reality, many of us do still go to shops, stores, malls, boutiques, supermarkets, and real-world outdoor markets all the time. The point is, the rise of online shopping can be traced back to pre-millennial times, and the global events associated with the pandemic have done nothing to stem the rise in clicks vs. bricks retailing.

The reality of the post Covid-19 era is that online shopping is happening everywhere, i.e. not just on laptops, tablets, smartphones, and other essentially desktop-based machines, but across a huge range of other different digital touchpoints. Every device from smart TVs to automobiles to airport kiosks can provide us with a digital shopping experience if it wants to… and it probably soon will.

The massive scope of the digital strip mall of the future presents us with an almost infinite variety of shopping conduits. This huge array of options actually makes choice more difficult, for humans at least. We can now look to machine intelligence to help us understand what we would most likely prefer to buy when, where, and why.

Recommendations required

What we (consumers) need now are better recommendations. 

Crucially though, we need those recommendations to be delivered across the variety of burgeoning digital touchpoints, so it makes sense to use an Artificial Intelligence (AI)-optimized Application Programming Interface (API) approach to allow recommendations to surface wherever we as consumers happen to be. 

Perhaps even more crucially, we need recommendations based upon our behavior and preferences surfaced within milliseconds… and we need those recommendations to be capable of drawing upon an entire range of product or service data that encompasses everything from description, availability, price, and onwards to aspects of compatibility and usability where appropriate.

According to Jordan Jewell, research manager, digital commerce at IDC: “Due to COVID-19, a record share of retail sales saw record growth in 2020, raising the stakes for practically every organization to have a digital commerce strategy. In this hyper-competitive market, merchants must provide customers with unique, personalized, and frictionless commerce experiences to succeed.”

Jewell agrees that an API-first tech stack is the foundation of these differentiated experiences. 

Cart expansion & average order value

Where the early e-commerce pioneers would have been happy with any customer who actually completed their product or service selection, filled in all required form-fields and subsequently executed payment successfully, modern e-business is far more demanding and exacting. Today’s e-commerce vendors need to be able to maximize the average order value through shopping cart expansion and customer satisfaction in their online stores. 

Significant progress towards weightier carts and orders can be achieved by using Algolia Recommend, a technology advancement that enables retailers to earn greater trust and loyalty by demonstrating a richer understanding of their customers by surfacing highly relevant recommendations in the  moment.

In a recent survey, 42% of respondents said that it was “very important” or “somewhat important” to see personalized content (such as recommendations, offers, or other customer’s previous experiences) when purchasing online. When studying the impact of product recommendations in the U.S., 38% of respondents stated they would shop “much more frequently” or “more frequently” at online retailers if they received such recommendations.

In an example of one use-case, European telecoms provider Orange România used Algolia Recommend technology to retain and convert shoppers when landing on out-of-stock products, an action that very typically might see shoppers abandon a page, entire website or even brand.

An API-first approach

Of course, it’s easier to talk about using a recommendation engine than it is to develop one from scratch, stress test it, debug it, integrate it and keep it maintained, managed and highly performant. This is why Algolia Recommend is delivered via an API-first approach that is simple to integrate and easy to use. 

The existing systems and software stack in any organization can be elevated to the advantages on offer here – and the functionality far outstrips off-the-shelf packaged solutions, which make it almost impossible to develop a differentiated customer experience or gain a competitive advantage. 

Algolia Recommend initially includes two of the more popular Machine Learning models that automatically deliver tailored recommendations. The “Related Products” recommendation model enables retailers to increase conversions and orders by analyzing items a shopper interacts with (e.g. clicks, adds to a cart, and/or purchases) and suggesting similar products during the same session.

The “Frequently Bought Together” recommendation model increases average order value by upselling complementary items on the product page or shopping cart page based on how other shoppers have interacted with that same item during a single shopping session.

Beyond the search box

As co-founder and chief technology officer of Algolia Julien Lemoine has said, this is all about helping customers to “go beyond the search box” and start experiencing an optimized online shopping experience that really drives increased revenue.  

“Algolia recently unveiled its new company direction and vision and helped customers go beyond the search box with their digital commerce strategies,” said Lemoine. “The release of Algolia Recommend provides the next building block for retailers to optimize their online experience and increase their revenue. These retailers have already unlocked $1 billion+ additional annual revenue on the back of up to 1.7 trillion searches across Algolia’s API platform.”

Just as the pre-Internet man or woman in the street might have “only” had the stores and the mall to go to, the pre-recommendation engine e-commerce shopper “only” had the search box to rely upon… today we have an ability to search smarter with API-driven machine intelligence designed to fill our carts and warm our hearts. 

 

About the authorAdrian Bridgwater

Adrian Bridgwater

Enterprise Software Industry Journalist

Recommended Articles

Powered by Algolia AI Recommendations

Introducing Algolia Recommend: The next best way for developers to increase revenue
Product

Introducing Algolia Recommend: The next best way for developers to increase revenue

Matthieu Blandineau

Matthieu Blandineau

Sr. Product Marketing Manager
What is a product recommender (or product recommendation engine)?
Product

What is a product recommender (or product recommendation engine)?

Catherine Dee

Catherine Dee

Search and Discovery writer
How any business can benefit from personalization and recommendations
Product

How any business can benefit from personalization and recommendations

Marie-Laure Thuret

Marie-Laure Thuret

Technical Product Manager